Mathematical models have contributed to many of the basic ideas leading to our current understanding of the urinary concentrating mechanism. However, these models have been unable to generate any significant concentration gradient in the inner medulla when experimentally measured parameter values are used. The long-term objective of this project is to develop comprehensive mathematical models for the urine concentrating mechanism in mammalian kidney. The specific aims of this pilot project are: to explore the typical experimental parameter space by formulating and numerically solving inverse problems for one-dimensional models; and to systematically evaluate and justify more rigorously the model simplifying assumptions by formulating models in two or three dimensions for isolated tubules, in which the simplifying assumptions of the one-dimensional models can be removed.